Sensor Alert & Performance(SAP)
This will give us an overview of the low-cost(Purple Air) sensors networks across India & specifically over Indo-Gangetic Plain. We can look into the AQ-Particluate matter of 2.5 microns (PM2.5) data based on the solutions (projects) like SAMOSA, NASA, UTN or by sensor names such as NASA_AQCS_102, UTN102 or SAMOSA_0001. We can further filter based on city, district or state. We can find the number of sensors under each project & also easy to download in .csv, .xlsx or .pdf file formats.
The map provides an interactive interface to view the distribution of low-cost sensors network across India. The AQI is based on average PM2.5 (Indian standards) in last 1 day. We can also click any sensor unit to get brief about the selected sensor.
This gives us the list of sensor units that are inactive for past 24 hours.
Purple Air sensors have two channels,Channel A & Channel B. This provides hourly aggregated R2 (coefficient of determination) and RMSE(Root Mean Square Error) among the 2 channels of each sensor unit for last 30 days. Observations provides the number of values for last 30 days for a sensor. The total number should equals 720 (Hour*days, [24*30]). We can easily filter based on the metric values to see how the sensors have performed over last 30 days.
Here we can explore & analyse the time series pattern of a sensor or multiple sensors for last 30 days. The time series measures PM2.5 concentrations of both channels A & B. Scatter plot helps us to know how each channel of the sensors are performing. We can compare against sensors, distcirts or states. In QA-QC, check box 0 is to see noisy data and 1 to see data after performing basic QA-QC. This will help us to diagnose the non-perfroming channel.
This allows to compare the PM2.5 levels of all sensor units across states and districts via Box plot . Double click (tap) on the state legend to explore the box plot for a specific state. Select multiple states for comparative analsyis of PM2.5 levels among states.
For any query or suggestions reach out at : adeel.khan@ceew.in